Message/Author |
|
Luwis Diya posted on Wednesday, June 23, 2010 - 11:25 am
|
|
|
I am trying to fit a Multilevel CFA with ordinal variables and I am always getting the message that the system is running out of memory. Are there ways to overcome this? |
|
|
It sounds like you have have several dimensions of integration. You can try using the WLSMV estimator instead of maximum likelihood which is the default. |
|
|
I am running a two-level CFA with 6 binary and 6 ordinal indicators and receiving this error: THE STANDARD ERRORS OF THE MODEL PARAMETER ESTIMATES COULD NOT BE COMPUTED. THE MODEL MAY NOT BE IDENTIFIED. CHECK YOUR MODEL. PROBLEM INVOLVING THE FOLLOWING PARAMETER: Parameter 64, %BETWEEN%: [ SCF$1 ] THE CONDITION NUMBER IS 0.114D-13. Is there any reason why the tow level model is not identifiable? This the model input. VARIABLE: NAMES = psu Sex randsex GMB CA CAD SSE SSF SSI TRUSTL TRUSTN TRUSTS SCR SCSB SCF; USEVARIABLES = GMB CA CAD SSE SSF SSI TRUSTL TRUSTN TRUSTS SCR SCSB SCF; CATEGORICAL = GMB CA CAD SSE SSF SSI TRUSTL TRUSTN TRUSTS SCR SCSB SCF; CLUSTER = psu; USEOBSERVATION = (Sex EQ 2); Missing are all (999); Analysis: TYPE = TWOLEVEL; Estimator = WLSM; MODEL: %WITHIN% fw1 BY GMB* CA CAD; fw2 BY SSE* SSF SSI; fw3 BY TRUSTL* TRUSTS TRUSTN; fw4 BY SCR* SCSB SCF; fw1@1 fw2@1 fw3@1 fw4@1; %BETWEEN% fb1 BY GMB* CA CAD; fb2 BY SSE* SSF SSI; fb3 BY TRUSTL* TRUSTS TRUSTN; fb4 BY SCR* SCSB SCF; fb1@1 fb2@1 fb3@1 fb4@1; |
|
|
Perhaps that threshold estimate has an extremely large value. If this doesn't help, send your output to Support along with your license number. |
|
Back to top |